[pve-devel] [PATCH http-server 1/3] accept-phase: fix conn_count "leak"

[Stoiko Ivanov]

When handling new connections in 'accept_connections' the number of
active connections got increased before the AnyEvent::Handle
registered the callback which would decrement it on error.

Any error/die beforehand would skip the decrement, and leave the
process in an endless loop upon exiting in wait_end_loop.

This can happen e.g. when the call to getpeername fails, or if the
connection is denied by the ALLOW_FROM/DENY_FROM settings in
'/etc/default/pveproxy' (which is also the simplest reproducer for
that).

Additionally it can cause a denial of service, by attempting to
connect from a denied ip until the connection count exeeds the maximum
connections of all child-processes.

Reported via our community-forum:
https://forum.proxmox.com/threads/pveproxy-eats-available-ram.79617/

Co-Authored-by: Dominik Csapak <d.csapak at proxmox.com>
Signed-off-by: Stoiko Ivanov <s.ivanov at proxmox.com>
---
 PVE/APIServer/AnyEvent.pm | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/PVE/APIServer/AnyEvent.pm b/PVE/APIServer/AnyEvent.pm
index c55da7f..c5f5fdc 100644
--- a/PVE/APIServer/AnyEvent.pm
+++ b/PVE/APIServer/AnyEvent.pm
@@ -1479,8 +1479,6 @@ sub accept {
 
     fh_nonblocking $clientfh, 1;
 
-    $self->{conn_count}++;
-
     return $clientfh;
 }
 
@@ -1561,7 +1559,7 @@ sub accept_connections {
 	    my $reqstate = { keep_alive => $self->{keep_alive} };
 
 	    # stop keep-alive when there are many open connections
-	    if ($self->{conn_count} >= $self->{max_conn_soft_limit}) {
+	    if ($self->{conn_count}+1 >= $self->{max_conn_soft_limit}) {
 		$reqstate->{keep_alive} = 0;
 	    }
 
@@ -1600,6 +1598,9 @@ sub accept_connections {
 		},
 		($self->{tls_ctx} ? (tls => "accept", tls_ctx => $self->{tls_ctx}) : ()));
 
+	    $self->{conn_count}++;
+
+
 	    print "$$: ACCEPT FH" .  $clientfh->fileno() . " CONN$self->{conn_count}\n" if $self->{debug};
 
 	    $self->push_request_header($reqstate);
-- 
2.20.1